Lake water levels fluctuate at time scales ranging from minutes to many years. While extensive research has been devoted to seasonal and multiannual time scales, we examined water level changes at time scales of minutes to hours. By applying impulse response functions to Lake Biwa and its catchment, we demonstrated that it is possible to (1) identify the effects of natural oscillations governed by the shape of the lake, (2) determine the precipitation retention time at the surface (above the impermeable layer) of the catchment, and (3) estimate how long it takes precipitation falling in catchment areas to reach the lake. Between 2010 and 2012 we recorded water levels at high frequency in the South Basin of Lake Biwa and calculated response functions between precipitation and water levels. We categorized the delay time determined from the calculated response functions as direct precipitation on the lake, seiches, and river inflow (surface flow, return inflow). We then identified the runoff patterns of surface and return flows by considering the time scale of the delay. The method used in this study enables flood predictions in cases involving heavy rain and appropriate planning of preventive measures against flood damage. These results will be valuable for the development of management strategies to respond to climate change in lake catchment systems.

湖泊水位在数分钟至数年的时间范围内波动。虽然针对季节和多年期的时间尺度进行了广泛的研究，但我们在几分钟至几小时的时间尺度上检查了水位的变化。通过将脉冲响应函数应用于琵琶湖及其流域，我们证明了可以（1）识别由湖泊形状控制的自然振荡的影响，（2）确定地表的降水保留时间（高于（3）估算降雨在集水区降落到达湖泊所需的时间。在2010年至2012年之间，我们记录了琵琶湖南部盆地高频水位，并计算了降水与水位之间的响应函数。我们将根据计算出的响应函数确定的延迟时间归类为湖泊上的直接降水，塞奇河和河流入流（地表流，回流）。然后，通过考虑延迟的时间尺度，我们确定了地表水和回流的径流模式。本研究中使用的方法能够在涉及大雨的情况下进行洪水预测，并适当规划预防洪水的措施。这些结果对于制定应对湖泊集水系统气候变化的管理策略将具有重要价值。本研究中使用的方法能够在涉及大雨的情况下进行洪水预测，并适当规划预防洪水的措施。这些结果对于制定应对湖泊集水系统气候变化的管理策略将具有重要价值。本研究中使用的方法能够在涉及大雨的情况下进行洪水预测，并适当计划预防洪水的措施。这些结果对于制定应对湖泊集水系统气候变化的管理策略将具有重要价值。